Visual lab / Engineering archive
Interactive demonstrations, architecture studies, and visual experiments from my engineering practice. Diagrams and simulated metrics illustrate concepts; they are not research evaluation results.
Metrics
Years Experience
Building production systems
Projects Shipped
Full-stack applications
Technologies
Languages & frameworks
Contributions
commits & deliveries
Selected work — flagship systems
Five systems, five different problems — retrieval, agents, defense operations, clinical AI, and lending. Select a system to inspect its architecture.
SYS-01 — Atomic Retrieval Fabric
The evidence layer for production AI. Sources become addressable atoms with locators — page boxes, table cells, JSON paths — never lossy chunks. Hybrid retrieval with RRF fusion assembles evidence packets, every claim is verified against its atoms, and a BLAKE3-hashed, Merkle-backed audit trail makes each answer replayable. Chunks are derived artifacts. Atoms are the truth.
What I do / Data · GenAI · Full-stack
I build by the evidence: understand the mathematics, test the assumptions, and inspect what fails. These are the decisions behind the tools.
An answer should lead back to the material that supports it. I build around addressable evidence: a page region, a table cell, a JSON path. Retrieval, generation, and verification are separate responsibilities.
BM25, dense embeddings, and learned sparse retrieval offer different signals. Fuse the rankings, rerank the candidates, and check each claim against its cited source. A citation is a starting point for verification, not proof by itself.
// Retrieve first. Evaluate support separately.let hits = rrf_fuse(&[ bm25.search(query), dense.search(query), sparse.search(query),]);let evidence = rerank(query, hits);let draft = generate(query, &evidence);match check_support(&draft, &evidence) { Supported => release(draft.with_citations()), Insufficient => abstain("missing evidence"),}# A useful result is a reproducible one.def recall_at_k(retrieved, relevant, k): if not relevant: return None # no labeled positives hits = set(retrieved[:k]) & set(relevant) return len(hits) / len(set(relevant)) # Report by query slice, not only an average.# Inspect misses and unsupported citations.# Keep the corpus and label versions fixed.Illustrative patterns. Simplified for reading; these are not public SDK contracts or performance results.
Interactive study / Retrieval
Pick a question. Inspect its sources. See where the answer has to stop. The useful part of retrieval is the connection you can check.
Scripted demonstration. Synthetic records and prewritten answers; no live model, database connection, or patient data. The clinical example illustrates data lineage, not medical advice.
Choose a question
The fourth question has no supported answer. That is part of the demonstration.
Where does the OSA severity flag come from?
Source atoms / Select to inspect
Sample answer / Inspect the citations
In this fictional record, the report states an AHI of 34.2 and a severe OSA finding . The table preserves that value and its source locator . The ETL copies the extracted finding into the severity field . This explains the record's lineage; it does not establish a clinical diagnosis.
The verification question
The report, stored value, and transform support the narrow claim about where this field came from.
Expertise / from source to system
I design and build across the stack: data pipelines, ML systems, backend services, frontend applications, and cloud infrastructure. My work carries through deployment, security, and production operations. Explore the disciplines behind that end-to-end practice.
End-to-end delivery
Ingest · model · transform
Train · retrieve · evaluate
Interfaces · APIs · services
Cloud · infrastructure · CI/CD
Observe · secure · improve
Engineering disciplines
07 connected areas of practiceHover or focus to explore. Click to pin; click again or press Escape to release.
01 / AI & machine learning
Retrieval, representation learning, and evaluation belong together. I work from a baseline toward hybrid retrieval, reranking, and claim-level checks; deep learning is useful when its assumptions fit the problem.
What I check
Inspect recall@k, citation faithfulness, and failure cases. A cited source still has to support the claim.
Architecture / Interfaces & invariants
The useful question is what happens between the boxes: who can act, what can fail, and how the result can be checked.
01 / Client
02 / Edge / gateway
03 / Services
04 / Data layer
05 / AI / ML
Inspecting / Services
Coordinate retrieval, models, and tools with bounded execution. Record evidence and evaluation decisions so failures can be investigated beyond the final text.
The design question
What can this service claim, and when must it abstain?
Hover or focus to trace relationships. Select a component to inspect its purpose, assumptions, and neighboring services.
Engineering philosophy
Aerospace engineering shaped how I think about software: understand the requirement, make the interfaces explicit, and test what happens when assumptions break.
For machine learning, that means looking beneath a score. What objective did we optimize? What information was available? Which errors matter? The mathematics explains the method; experiments establish its limits.
The experience behind the practiceRequirement
What must hold?
Model
Under which assumptions?
Evidence
How would we know?
01 / Working principle
Define the operating conditions, failure modes, and acceptance criteria before choosing an architecture.
02 / Working principle
Start with the mathematics or mechanism. Record assumptions, units, uncertainty, and where the model stops being useful.
03 / Working principle
Compare against a baseline. Test boundaries, inspect errors, and keep enough evidence for someone else to repeat the check.
Technologies / working inventory
Chosen for the work: a query engine for the data, a runtime for the workload, an interface for the person using it. The interesting part is how the pieces meet.
01
Express the model. Make the result usable.
02
Represent, retrieve, evaluate, and serve.
03
Move data with its meaning intact.
04
Control concurrency and define boundaries.
05
Provision, release, observe, recover.
06
Make permissions and evidence inspectable.
Selected systems / Applied engineering
Retrieval, governed development, and computing at the disconnected edge. These technical write-ups describe the implementation choices behind Structura, Praxis, and AegisOS.
Atomic Retrieval Fabric
The evidence layer for production AI — sources stay addressable as atoms, every claim is verified, and BLAKE3 + Merkle proofs make each answer replayable.
Evidence-Native Dev Platform
Learns a business, builds a governed knowledge fabric and graph, stands up grounded agents, and runs a software factory that ships reviewable code.
Offline-First Secure Edge OS
A Rust-first AI operating system for the disconnected edge — cryptographic node identity, tamper-evident timeline, encrypted vault, multi-path mesh, offline RAG.
Further engineering work
Clinical retrieval pipeline
Retrieval-augmented pipeline over clinical documents — semantic chunking, hybrid retrieval with reranking, and LLM synthesis under PHI guardrails with enforced citations.
Full-stack lending system
A lending workflow connecting application intake, document verification, risk assessment, and underwriting review.
Data & visualization tooling
Visual data tooling for engineering teams — turning raw pipeline output into explorable, decision-ready views.
Signal & data processing
Processing infrastructure for high-volume signal data — ingestion, transformation, and analysis built for throughput.
AI monitoring system
Intelligent monitoring — anomaly detection and alerting over live operational data streams.
Security tooling
Defense-minded security tooling — policy, visibility, and enforcement primitives for hardened environments.
Journey
2019
Unity Engine w/Dev
Started coding games with Unity AR/VR
2021
Lex Fusion & Credora
CTO — led full-stack product engineering & DevOps
2022
IBM Watson Platform
Built AI automation on IBM Cloud via partner program
2023
ViDaT & SignalForge
Aviation data platform & RF anomaly detection for defense
2024
Lockheed Martin SBIR
Defense contracts, SBIR proposals & POCs via supplier portal
2024
Aviation Data Engineering
Supporting mission-critical logistics and sustainment programs
2025
SteerBridge — F-35
Fleet-readiness data pipelines & DoD analytics for the F-35 program
2025
MDC · Guardis · Relay
Veterans' claims pipeline, civic AI safety & gov API gateway
2025-26
Structura · Praxis · Aegis
Evidence-grade AI, agentic platforms & secure edge systems
May 2026 - Present
Accenture Federal Services
Engineering agentic and financial systems for federal mission programs
2019
Unity Engine w/Dev
Started coding games with Unity AR/VR
2021
Lex Fusion & Credora
CTO — led full-stack product engineering & DevOps
2022
IBM Watson Platform
Built AI automation on IBM Cloud via partner program
2023
ViDaT & SignalForge
Aviation data platform & RF anomaly detection for defense
2024
Lockheed Martin SBIR
Defense contracts, SBIR proposals & POCs via supplier portal
2024
Aviation Data Engineering
Supporting mission-critical logistics and sustainment programs
2025
SteerBridge — F-35
Fleet-readiness data pipelines & DoD analytics for the F-35 program
2025
MDC · Guardis · Relay
Veterans' claims pipeline, civic AI safety & gov API gateway
2025-26
Structura · Praxis · Aegis
Evidence-grade AI, agentic platforms & secure edge systems
May 2026 - Present
Accenture Federal Services
Engineering agentic and financial systems for federal mission programs
Blog
I'm an AI/ML, data, and LLM engineer who builds systems that have to be right the first time — evidence-grade retrieval, agentic platforms, and mission-critical data infrastructure. From F-35 fleet-readiness pipelines to my own flagship systems — Structura, Praxis, and AegisOS — I bring a systems-thinking approach to every layer.
When I'm not shipping, I write about retrieval, LLM engineering, and secure systems — and build AI that proves what it claims instead of guessing.
Learn moreTechnologies
Let's build
Evidence-grade AI, mission-critical infrastructure, or a platform that needs to be right the first time — if it has to hold weight, let's talk.
